Prasanga Dhungel

Master's in Informatics at TUM, Lifelong learner.

prof_pic.jpg

Hi, I am Prasanga, a Software Engineer with a passion for Machine Learning, particularly Robust and Interpretable ML, Machine Learning with Graphs, Low resource NLP, and Generative Modeling.

Currently, I am pursuing a Masters degree in Informatics at TUM, with a focus on Machine Learning, Analytics, Computer Vision, and High Performance Computing. I am parallely working as a Data Scientist at E.ON Digital Technology. In past, I worked as a Data Scientist at Naamche where I was focused on leveraging large-scale real estate data to assess investment potential in the industry. Prior to that, I was a NLP Engineer at Diyo.AI. I received my undergraduate degree in Computer Engineering from the Institute of Engineering, Pulchowk Campus at Tribhuvan University.

Research Imperatives

The following is a list of areas that I have conducted research on or would like to work on in the near future:

  • Explainable AI (XAI): To facilitate learning and satisfy curiosity about why certain inferences and behaviors of models are such, interpretability and explanations are crucial. I believe that interpretability will improve ML models, make them more transparent and thus trustworthy. Precisely, I am interested in model-agnostic methods to explain AI models, so that the same techniques are applicable for the interpretation of all the models.

  • Low-resource NLP: In a low-resource settings like that of Nepal, where labelled data are particularly scarce, techniques like transfer-learning and self supervised learning can be tapped to ameliorate NLP components such as NER, POS-taggings, dependency parsing and many others. My interest in this area stems from the desire to bring powerful technologies like chatbots and voicebots – which are currently only available in resource-rich languages – to a wider audience.

  • Computer Vision in Medical Imaging: My work on medical image segmentation at NAAMII has taught me that computer vision can be a great tool for helping radiologist and surgeons find lesions, tumors, and fractures. This experience has fueled my passion for using computer vision to enhance the healthcare industry.

  • Algorithm Optimization: There are numerous exciting algorithmic and optimization challenges in various product areas. These include optimizing internal systems such as scheduling computational machines, as well as optimizing core products and user experiences through techniques like graph clustering and finding optimal paths in transportation networks. I am eager to contribute new algorithmic ideas that have the potential to impact millions of people around the world.

If you are interested in collaborating with me on any research projects or have ideas for future endeavors, please feel free to reach out. I am always open to discussing new ideas and ways to work together to create something amazing.


Miscellaneous

I am a lover of literature, intellectual podcasts, and the beauty of nature. In my free time, you can find me lost in a good book, exploring the great outdoors, or indulging in a thought-provoking film. But my true passion lies in the ability to convey complex ideas in a clear and accessible manner through writing. I am excited to share my thoughts and perspectives with you through this platform. Thank you for visiting!